[Rework] URL: Another update for urls extraction logic

[Rework] URL: Another update for urls extraction logic

URL extraction from HTML parts should look like this:
1. Extract href links
2. Convert HTML to plain text and extract:
  a) (http|https|ftp)://foo.bar and www.foo
  b) email like strings \bfoo@bar.baz\b .
For all extracted strings check if we have host with a domain from the public suffix.